基于脉冲噪声检测与信号压缩的滤波算法

针对传统滤波算法细节保护能力不强的缺点，提出一种基于脉冲噪声检测与信号压缩的滤波算法。在噪声检测阶段排序检测窗口内的像素点，判定极值附近灰度值落在一定范围内的点为噪声点，并生成噪声模版；在噪声滤除阶段提取滤波窗口内信息点并压缩，进一步排除误检。根据不同噪声密度进行不同滤波，并循环迭代直至整个图像中的所有脉冲噪声点被滤除。实验结果表明，该算法具有较好的噪声滤除和细节保护能力，尤其是在高强度噪声条件下效果更明显。

Filtering Algorithm Based on Impulse Noise Detection and Signal Compression

Aiming at the problem that the ability of details protection is not well in traditional algorithms, this paper proposes a filtering algorithm based on impulse noise detection and signal compression. During noise detection, it sorts the pixel points in detection window, and the pixel points in the field of extreme value in the vicinity whose gray value fell within a certain range are judged to be noise points. During noise filtering, it extracts signal points and compresses these points so as to further exclude false detections. It chooses different filtering method according to different noise density, and does cyclic iteration until all impulse noise points are filtered. Experimental results show that the algorithm has a better ability of filtering impulse noise and protecting details, especially under the conditions of high-intensity noise, it has more obvious effect.